Measures are a simplification of the complex, multifaceted nature of real life into a data point. They are a pauperised, context-free substitute for reality.
A single statistic conveys very little meaningful insight, unless combined with other metrics. In other words, the raw materials must be repackaged into a functional finished product.
Definition
In physics, a measure is a non-negative expansive property of a manifold or an object. For instance, the gravitational potential is a measure. Also, the Liouville and Gibbs measures are examples of measures.
A measure is also a segment of music defined by a certain number of beats and separated by vertical bar lines. It helps musicians keep a steady tempo and provides structure to the composition.
Another important distinction between metrics and measures is that measurements focus on inputs such as resources and activities, while metrics track progress toward desired outputs based on those inputs. For example, when a product is measured with a ruler, the result is a value that can be used to make a decision about whether the product is acceptable or not. This determination is made based on inspection, which compares the measurement against available references (in this case, a standard ruler). Metrics use data over time to make predictions about future performance.
Purpose
Measures and metrics are useful tools to identify areas for improvement. However, they are useless unless the data you collect is accurate and aligned with your goals. Metrics are more focused on inputs and provide quantitative evidence of progress toward a goal, while measures focus on the result and how to make improvements.
The best measures are those that are clearly linked to strategic goals and priorities. They should be crafted at the same time as the goals themselves. If a measurement does not support your goals, consider dropping it.
Meaningful measurement involves understanding people – those who will be measured and the ones who will use the data. Clearly communicating purpose and impact in ways that matter helps everyone build confidence in the process and trust in its outcomes. Tip: Make your measures easy to understand. Vague ideas, surveys and kooky acronyms are not a good start. Consider involving the people who will respond and inviting them to a Measure Gallery to learn about your measures before they take part.
Implementation
The measure implementation process encompasses the activities required to progress a quality measure from its development state into an active, in-use state. It includes establishing quality criteria for measurement, promoting the use of the measures, and assessing their effectiveness in improving healthcare practices. It also involves addressing any potential biases that may be inherent in the measures, such as those that might disproportionately affect certain populations.
A key step in the implementation process is identifying pragmatic measures that are relevant to stakeholders and feasible for use in practice. A previous review found that a large percentage of measures had unknown psychometric qualities and pragmatic strengths, making it difficult for researchers and practitioners to select pragmatic measures. The current study aims to address these gaps by developing a methodology for systematically evaluating pragmatic measures and establishing a set of pragmatic rating criteria. The resulting ratings will enable the field to identify and select pragmatic measures that are both psychometrically strong and pragmatic.
Monitoring
Monitoring is an ongoing process that provides valuable insights into a system’s health. It identifies areas for improvement and highlights the impact of specific policies. It also allows for quick response and mitigation of risks, enhancing risk management, operational resilience, and regulatory compliance.
It can be conducted at different levels, including project, programme, organisation and sector. It can also be facilitated by partnerships across organisations, countries and/or sectors. Monitoring can be time-consuming, especially if it involves collecting qualitative data and documenting lessons learned.
Observability complements monitoring by adding context to the data. While monitoring is goal-driven and encourages teams to track known metrics, observability goes beyond this to detect anomalies in the data, identify root causes, and enable preventive actions. Both tools are important to building resilient applications, especially in CI/CD workflows. However, it is essential to understand the difference between them. Using the wrong tool can lead to false alarms and inaccurate results. This could impact your business’s bottom line and reputation.